Z15 - Modeling Fibrosis in a Collagen Mimetic PEG Scaffold

Introduction: : Liver fibrosis progresses through tissue injury, resulting in a complex cascade of environmental signals, such as cell-to-cell, biomolecular, and mechanical signals, that trans-differentiate hepatic stellate cells (HSCs) into myofibroblasts. This leads to an increased production of extracellular matrix (ECM) and an increased tissue stiffness that acts as a positive feedback loop that causes additional injury and may eventually result in cirrhosis and an increased risk of liver failure and cancer. Despite advances in treatments, many options remain ineffective. By understanding the underlying mechanisms involved in this feedback loop, we can further develop diagnostic tools and therapeutics. Previously, researchers approached this challenge by creating 2D in vitro models that were limited in their ability to accurately recreate the in vivo environment and the corresponding signals. Recently, 3D models have shown promise in modeling environmental cues for fibroblasts, however very few of these studies have been done with HSCs. Likewise, no study has investigated the effect of the topological cues from the native ECM fibers on HSC activation. To overcome these limitations, we’ve interfaced fibroblasts with a hydrogel system that can be independently stiffness-tuned, functionalized with ECM-binding motifs, and crosslinked with a collagen-mimetic peptide (CMP) that mimics the physical structure of collagen I fibers. With this foundation, we will interface iPSC-derived HSCs to further model the in vivo disease. This approach allows us to independently tune biomolecular, mechanical, and microphysiological signals to elucidate new mechanisms in liver fibrosis pathogenesis so that new strategies for treatment and therapeutics can be discovered.

Materials and

Methods: : We first established a baseline of fibroblast activation in 2D by culturing fibroblasts on plastic cell culture dishes and activating them with TGFβ, a known activator of both fibroblasts and HSCs. To quantify the activation, we measured the difference in activation markers, alpha smooth muscle actin (αSMA) and collagen I, via immunofluorescence microscopy. To create the hydrogel model, we utilized a stiffness-tunable gel with a poly(ethylene glycol) backbone crosslinked with a cell-degradable crosslinker. We then functionalized the gels with either RGD or GFOGER ECM-binding motifs. Likewise, we functionalized the gels either with or without the CMP. To determine the effect of these functionalizations, we embedded cells into the gels and imaged them with phase contrast microscopy and quantified the changes in the percentage of cells spreading and the average cell spreading areas. Furthermore, we treated the cells with TGFβ and measured the difference in the activation markers, alpha smooth muscle actin and collagen I, through immunofluorescence microscopy. Towards the goal of incorporating HSCs into the gel system, we have adapted a protocol from literature to derive HSCs from iPSCs. We then validated the iPSC-HSCs for known HSC markers (αSMA, Vimentin, GFAP, PGDFRβ, and PCDH7) through immunofluorescence microscopy and qRT-PCR. Furthermore, functional assessment was performed through measuring vitamin A storage through 2-photon confocal autofluorescence imaging.

Results, Conclusions, and Discussions:: Fibroblasts cultured in 2D exhibit a typical spindle-like morphology. Once treated with TGFβ, the cells exhibit a trans-differentiation into myofibroblasts, as seen by the morphology change in Figure 1A. Additionally, the cells exhibit a 3-fold increase in αSMA and incorporation of αSMA into their cytoskeletons and a 6-fold increase in collagen I expression, as shown in Figure 1A. Together, these results demonstrate a transition of fibroblasts into myofibroblasts when the cells are exposed to TGFβ. In the 3D hydrogel system, fibroblasts did not exhibit any spreading when the matrix lacked a binding motif, as shown in Figure 1B. In contrast, the cells showed an increase in cell spreading and percentage of cells spreading in gels that were functionalized with RGD in comparison to those functionalized with GFOGER. While both binding motifs would be present in vivo, these findings highlight the differential response of these cells to various environmental cues. The iPSC-HSCs were assessed for their ability to store Vitamin A, a key hallmark of HSCs. In contrast to primary cells and HSC cell lines, our iPSC-HSCs showed the ability to store Vitamin A, as depicted in Figure 1C. Furthermore, the cells were characterized for HSC markers and were found to be positive for the markers PGDFRβ, Vimentin, GFAP, and αSMA). Likewise, the cells were positive for PCDH7 and showed nearly an 80-fold increase in the marker over the course of the differentiation. Together, these results validate our iPSC-HSCs and show that they are positioned in a more quiescent state than primary cells and cell lines for HSCs. From these results, we have designed a system to model liver fibrosis by recreating the disease in vitro with the ability to independently investigate the effect of biomolecular, mechanical, and microphysiological cues independently on the progression of liver fibrosis and we have paved the way for developing a model that further recreates the in vivo environment with HSCs.

Acknowledgements and/or References (Optional)::